Kimi k2.6/7 running inside Kimi code already kicks the pants of the latest Claude and OpenAI models when it comes to cyber security. I regularly run multi model security reviews and while opus 4.6/7/8 and gpt 5.3/4/5 find a couple of things and declare mission accomplished (running inside pi) kimi k2.6/7 inside pi finds more issues and inside kimi code finds the most.

There are sometimes false positives but when I give Kimi’s report to the frontier models they more often than not confirm they are valid security issues but didn’t find them themselves.

> So although it's nice to have capable open models, it will increasingly bad for us all if open models keep fast-following closed model capabilities as they have been

Cat's out of the bag. The only way to make them safe is to make sure everyone has access to them. This might be an iffy analogy, but if Dario uses it all the time then so can I: they're kinda like nuclear weapons. If only one country has access to nukes then you're in trouble. If everyone has access to them, then it's mutually assured destruction to use them.

Sure, it could be increasingly bad if open models keep increasing in capability. But it will be much, much worse if only the rich and the powerful have access to this technology, and us -- the have-nots -- will have to contend with whatever scraps we'll be allowed to eat off the table of whichever billionaire is in control. We've already seen a prelude of this with Mythos being restricted and Fable being suddenly yanked. Is this the world you want to live in? Where only Dario and his friends have access?